get_aws_instance_monitoring
Retrieve AWS EC2 instance monitoring metrics for specified time periods to analyze performance and track resource utilization.
Instructions
获取AWS EC2实例的监控数据
Args:
instance_id (str): EC2实例ID
hours (int): 获取过去多少小时的数据
Returns:
Dict: 监控数据Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| instance_id | Yes | ||
| hours | No |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |
Implementation Reference
- main.py:354-367 (handler)MCP tool registration and handler for 'get_aws_instance_monitoring'. This thin wrapper delegates the actual logic to the aws_provider instance.
@mcp.tool() def get_aws_instance_monitoring(instance_id: str, hours: int = 1) -> Dict: """ 获取AWS EC2实例的监控数据 Args: instance_id (str): EC2实例ID hours (int): 获取过去多少小时的数据 Returns: Dict: 监控数据 """ return aws_provider.get_instance_monitoring_data(instance_id, hours) - providers/aws_provider.py:274-342 (helper)Core implementation in AWSProvider class that fetches CloudWatch metrics (CPU, Network, Disk) for the given instance over the specified hours using boto3.
def get_instance_monitoring_data(self, instance_id: str, hours: int = 1) -> Dict: """ 获取实例的监控数据 Args: instance_id (str): EC2实例ID hours (int): 获取过去多少小时的数据 Returns: Dict: 监控数据或错误信息 """ if not self.available: return { 'error': f'AWS服务不可用: {getattr(self, "error", "未知错误")}', 'provider': 'aws' } try: end_time = datetime.utcnow() start_time = end_time - timedelta(hours=hours) metrics = ['CPUUtilization', 'NetworkIn', 'NetworkOut', 'DiskReadOps', 'DiskWriteOps'] monitoring_data = {} for metric in metrics: try: response = self.cloudwatch.get_metric_statistics( Namespace='AWS/EC2', MetricName=metric, Dimensions=[ { 'Name': 'InstanceId', 'Value': instance_id } ], StartTime=start_time, EndTime=end_time, Period=300, # 5分钟 Statistics=['Average', 'Maximum'] ) monitoring_data[metric] = { 'datapoints': len(response['Datapoints']), 'data': response['Datapoints'] } except Exception as e: monitoring_data[metric] = { 'error': str(e), 'datapoints': 0 } return { 'provider': 'aws', 'instance_id': instance_id, 'time_range': f'{hours}小时', 'metrics': monitoring_data } except ClientError as e: return { 'error': f'AWS API调用失败: {str(e)}', 'provider': 'aws' } except Exception as e: return { 'error': f'获取监控数据时发生错误: {str(e)}', 'provider': 'aws' }